Introduction

The starting point of the present article is rooted in the growing concern of lecturers in the Bachelor’s Degree in Primary Education about the difficulties of their students, i.e. future teachers, to use basic skills in reading comprehension and written expression for the construction of knowledge and psychoeducation. This situation has been already shown in previous studies (^{HERRADA; QUINTERO, 2012}; ^{ARROYO; JIMÉNEZ-BAENA, 2016} ; ^{DUGAREVA; DIEGUEZ; PÉREZ, 2014}; ^{RELLO; MERCADER; SIEGENTHALER; PRESENTATION, 2018}), as well as in PISA or PIAAC surveys: the Programme for International Student Assessment (MECD, 2015) and Program for the International Assessment of Adult Competencies (MECD, 2013) respectively and show a gloomy picture of the future because of the requirements of the students to excel in competencies to become knowledge builders, especially in the university context.

Considering the abovementioned aspects, this study has a double purpose. Firstly, the analysis on the competencies in reading comprehension and written expression of students accessing to the Bachelor’s Degree in Primary Education, investigating if those students who have faced previous university and academic learning and tasks, as those who have accessed to the University Degree in Psychopedagogy, show some improved command on the strategies involved in text comprehension and expression. And secondly, the consideration of the increasing influence of Information and Communication Technologies (ICTs) on learning environments and, consequently, the analysis on whether there are remarkable differences according to the stimulus presented when applying the abovementioned strategies in an academic task of knowledge access. This stimulus was the presentation of a printed text and a hyperlinked text, and the task consisted of generating a summary after reading both types of texts.

Based on this double purpose, the research establishes a series of aims resulting in different hypotheses. The objectives are as follows:

Theoretical framework

Different authors have highlighted the fact that communicative competencies, including written expression and spelling skills, are not assumed enough by university lecturers ( ^{SANELEUTERIO, 2018} ). For this reason, the author points out the needed adaptation of the curriculum to improve teacher training in these areas. Being in unison with this author, it is essential to analyze whether these students are skilled enough in reading comprehension and written expression, as these competences are fundamental to train their students in the future. In addition, it is not only to train future lecturers in reading comprehension and written expression techniques, but it is also necessary that they shall be able to evaluate these skills too (^{RUIZ; VÁZQUEZ; SEVILLANO, 2017}).

Before approaching the research, some key aspects shall be described; such as the role of microstructure, macrostructure and superstructure in reading comprehension; the -need to use techniques that allow the reader to manage their cognitive resources better during the reading process; and the description and analysis of the synthesis strategies used by mature readers to summarize texts.

Models of text comprehension

Text comprehension is the result of the interaction between the explicit text expressions and the reader knowledge. Some authors have analyzed how mental constructions depend as much on individuals’ representation of textual information as on the reader’s prior knowledge ( ^{TIJERO, 2009} ). Among the best known models we find the one presented by ^{Van Dijk and Kintsch (1983)} , and it states that the readers process the text information founded on three levels of representation: surface code, text base, and situation model. Subsequently, ^{Kintsch (1988)} proposed the construction-integration model as the readers construct two types of mental representation during the reading process: the textual representation or base, which is close to the text meaning, and the situational representation, which refers to the context the textual content is framed. In recent years we can highlight models such as the MD-TRACE proposed by ^{ROUET and BRITT (2011)} , which assume that readers use internal resources, such as prior knowledge of the subject matter of the text, the mental representation of the document’s content; external resources, such as external task specifications, the set of sources and documents used; and the products generated by readers, such as summaries or concept maps.

The function of textual microstructure and superstructure in the extraction of the macrostructure

The reader, as s/he reads a text, extracts predicative propositions or outlines formed by a verb and at least two arguments; and s/he also determines the degree of existing coherence between them at a local level. This brings on the elaboration of the microstructure. Thus, it is essential to recognize or infer the two types of relations, thematic and structural, that occur between a proposition and the previous and subsequent proposition ( ^{VAN DIJK, 2006} ).

Apart from identifying propositions and determining the degree of coherence among them, the reader must recognize which of those propositions are textually relevant and establish the coherence among them globally. It requires accessing to a deeper semantic level, called macrostructure, in order to identify and hierarchically reorganize the essential ideas ( ^{ESCORIZA, 2006} ; ^{VAN DIJK, 2006} ). Finally, the textual superstructure or organizational form of the text gains a fundamental role as it provides some keys about which microstructure propositions could form the macrostructure; facilitating, in addition, a scheme, adapted to the different textual typologies (problem-solution, causal, comparative, argumentative, etc.), to store the essential ideas in an organized and hierarchical way.

Cognitive management of the reading process

When the reader reads, does not process the text information at once, but in several parts or processing cycles that retain a reduced number of propositions for a limited time. Taking this as a reference, ^{Sánchez and Solé (2009)} and Sánchez, García and Rosales (2009, 2010) consider that a text can hardly be processed at a local and global level simultaneously. According to this, McNamara, Ozuru, Best and O’Reilly (2007) state that in order to understand complex texts it is necessary to simplify the task of reading, starting from self-regulating strategies such as re-reading, underlining and taking notes in the text, constructing conceptual maps and summarizing. Different authors have analyzed the use of metacognitive approaches to improve reading comprehension levels (^{RAMÍREZ; ROSSEL; NAZAR, 2015}; ^{CABERO-ALMENARA; PIÑERO; REYES, 2018}). Different studies, such as the one presented by ^{Vega, Torres and Del Campo (2017)}, indicate that morphological processing plays a central role in developing reading skills.

Synthesis strategies used by mature and immature readers

There are remarkable differences in the synthesis strategies used by mature and immature readers. Thus, immature readers tend to have comprehension problems both locally and globally, and therefore elaborate a composition closely linked to the surface code of the original text, using omission-selection strategies such as the following: copying (^{MANDIN; LEMAIRE; DESSUS, 2007}); paraphrasing, almost copying ( ^{HIRVELA; DU, 2013} ); and the linked combination, which involves the unstructured integration of two copied or paraphrased sentences or more from the original text. These difficulties in reading comprehension, moreover, often lead this type of reader to construct/infer ideas based on a misinterpretation of the textual content ( ^{MANDIN et al., 2007} ).

Mature readers, on the other hand, have no difficulty in understanding the text locally and globally, so they produce a summary closely linked to the model of the situation, generating a mental model that allows them to integrate the textual base into their previous knowledge. This model is constructed using strategies such as omission-selection, paraphrasing personally, generalization and construction (WESTBY; CULATTA; LAWRENCE; HALL-KENYON, 2010).

The generalization of the use of ICTs has meant that the level of maturity of readers may differ depending on the text format they use. In this sense, some recent studies analyze the reading comprehension using electronic devices. Thus, ^{Martínez and Esquivel (2017)} analyzed the existing differences in reading comprehension between students using multimedia resources and those reading printed texts: it was found that those students using multimedia resources obtained higher advances than the rest of them. In addition, other authors have demonstrated that multimedia programs facilitate reading comprehension for those students with difficulties (^{CABERO-ALMENARA; PIÑERO; REYES, 2018}).

Methodology

A set of hypotheses is raised based on the objectives described above:

In order to contrast the hypotheses, two studies are carried out. The first tries to answer the first three hypotheses, while the second focuses on the fourth. Both are framed within different designs that we will describe independently below.

Experimental design

For the first study ( Design I) we established a transversal design with a population with students and graduates of the Bachelor’s Degree in Primary Education at the University of Salamanca (academic year 2013/2014), in order to select two groups that were at different formative stages:

Primary Education Group ( PED ): students enrolled in the first year of Primary Education Studies at the University of Salamanca.

Psychopedagogy Group ( PSP ): students in the first year of Psychopedagogy after the completion of the Bachelor’s Degree in Primary Education during the same academic year.

We used a pre-experimental design where pre- and post-treatment measures of the dependent variable were not taken within the same study group; but each group represented a different measure by representing two different formative moments. Thus, we considered PED as the Pre-test measure; the formative process before being PSP as the treatment or Test; and PSP as the Post-test measure.

For the second study (Design II) we established a transversal design, based on the samples of the previous design, in order to form four groups made up of students in a specific formative moment and who had carried out the task of summarizing.

Thus, we used a quasi-experimental methodology with multiple observations of the dependent variable in each of the study groups to replace the control. In order to do this, we took as the Pre-test measure the results in the Design I test obtained by the students who composed the sample of this second design; as Test or treatment, the stimulus of the textual format (written or hyperlinked) when summarizing; and the results obtained in carrying out this task, as a Post-test measure.

Population and sample

The difficulties found in extracting randomly a sample from the study population led us to take three incidental groups as an initial sample: two from Primary Education and one from Psychopedagogy. We had access to them during the first three weeks of the academic year 2013/2014.

As described in Table 1 , the sample for Design I consisted of a total of 130 students, 96 of which participated in the tests exclusively corresponding to Design II. In other words, if we take into account that the design of study I is the Pre-test phase of Design II, once the corresponding tests have been carried out, the sample of this second design was made up of 96 students divided into four groups according to the formative moment they were, and the textual format in which they carried out the tests.

Instruments

Table 2 shows the different instruments used to measure the variables considered in the research, as well as the study and the phases these instruments were applied. In the first study we used:

In the second study, for the Pre-test phase, the instruments from the previous study were used to determine the students’ basic comprehension and written expression skills, and to analyze the considered control variables. In the Post-test phase, instruments were used to test whether students were capable of use strategically their comprehension and written expression skills to perform an academic simple task: summarize a document in a printed or hyperlinked format.

In this phase in particular, the dependent variable Specific skills of comprehension and written expression for the synthesis of information (VDTH) was analyzed through two operative variables, using the following instruments:

All the instruments were validated using the expert judgment technique, which included processes of content analysis individually by each expert, and also in groups, so that researchers could check the degree of coincidence and divergence in their evaluations, and discuss them until a high level of consensus was achieved.

Data analysis

Data analysis was carried out in two phases. Firstly, a group of experts analyzed and evaluated the content of the tasks carried out by students, first individually and then in groups, until a high degree of agreement was reached in their judgments. And secondly, different types of descriptive (contingency tables) and inferential (statistical χ², t-test of mean difference, ANOVA of a factor) statistical analyses were carried out.

Results

Discussion and conclusions

The first conclusion from the results obtained is that the formative moment in the university environment is not a determining factor for the possession of basic skills of reading comprehension and written expression in the sample considered. The analysis of the tasks of extraction of key ideas and construction of a conceptual map indicates that both the students who began the Bachelor’s Degree in Primary Education, and those who had accessed to the University Degree in Psychopedagogy (second cycle) from the Bachelor’s Degree in Primary Education, showed important deficits in reading skills such as following the thematic progression of the text, strategically using the textual superstructure to process it globally and self-regulating metacomprehension.

Secondly, textual format is not also a determining variable for predicting a better level of performance when summarizing an academic document in writing. The students, in general, constructed summaries with no clear global meaning, showing deficiencies that indicated the use of synthesis strategies more linked to the surface code of the text than to the construction of the situation model. Furthermore, taking into account that most of the students did not use a draft, it was difficult for them to manage their cognitive resources properly in order to process the text globally. All this seems to indicate that the text/hypertext structure was not strategically used to extract the macro-structure. On the other hand, more specifically, we see that those who read the academic document in printed format did not use techniques such as note-taking, a strategy that would have allowed them to highlight and synthesize relevant information. Although the most used technique was the underlining, only a few of them focused on aspects related to the structure of the text. However, both the students who read the academic document in hyperlinked format and those who read it in printed format did not use the intertextual information in the summary they wrote. If we take into account that the students in general were not able to use the superstructure strategically to extract the macrostructure from the text/hypertext, we consider that the absence of intertextual information in the summaries could be due more to comprehension difficulties than to the use of structural criteria.

According to different authors ( ^{MCNAMARA et al., 2007} ; ^{MEYER; RAY, 2011} ; ^{SALMERÓN; GUTIÉRREZ-BRAOJOS, 2012}) mature readers follow a series of procedures: goal setting, textual exploration or reading through in the before-reading stage; taking notes in the text, asking self questions and solving comprehension problems during the reading stage; and drafting in the after-reading stage. Considering all this, as well as the level of performance in the task, it seems evident that the procedures the students used to summarize were not adequate. None of them mentioned setting goals, taking notes in the text, asking oneself questions, solving comprehension problems, or drafting.

Despite this, and considering the differences found (LS=0.05), mainly linked to the formative moment, it is possible to affirm that, in certain aspects, the procedures used by the students of Psychopedagogy were better than those of Primary Education. The process used by the students of Psychopedagogy to summarize with the underlined information resulted more adequate than the one followed by the group of Primary Education, who carried it out directly from the reading of the text. The Psychopedagogy students read at least part of the hyperlinks, while a significant proportion of those of Primary Education did not access any of them. A good part of the Psychopedagogy sample used the glossary, while the majority of the students of Primary Education did not do so despite of the fact that they should be less familiar with the terminology. In general, the students stated that the summary task was very useful, and considered that they had no difficulties in carrying it out. This suggests that, although they knew the importance of the task, the students were not aware of the possible difficulties involved in its elaboration.

In short, it does not seem that the students who had completed the Bachelor’s Degree and were studying the University Degree in Psychopedagogy had acquired, after three years of academic training, effective skills in reading comprehension and written expression. On the contrary: they showed similar difficulties to the students enrolled in the first year of the Bachelor’s Degree in Primary Education when they had to face tasks such as extracting key ideas, constructing a conceptual map, or preparing a summary. With the results obtained, it is necessary for the educational community in general and educational agents with legislative capacity in particular to carry out an in-depth reflection on the measures to reduce the deficiencies of reading comprehension and written expression in university students. It is worthy to meditate whether educational policies, including universities’ curricula, should be adapted to improve training in these skills. It is paradoxical that certain publications, such as the PISA reports on reading comprehension of 15-year-old students generate great interest and even controversy internationally in the educational communities, whilst it is not usual to find studies on the competence of university students who, as those enrolled in the Bachelor’s Degree in Primary Education, will carry out their future work as teachers.

It is true that the results obtained at the University of Salamanca cannot be extended to other contexts due to the characteristics of the sample analyzed. However, we consider essential that, both in the initial formative stage (Bachelor’s Degree in Primary Education) and in the performance of their professional work, teachers shall receive adequate training in summary writing techniques, extraction of key ideas from texts and hypertexts, as well as in the design of concept maps. To this end, we estimate as necessary, in the first place, to develop programs for students enrolled in Bachelor’s Degrees in Primary Education, so that future teachers learn to use all the abovementioned strategies. These specific training programs must be integrated in the curriculum, and shall be related to communication and technology contents, where the two textual formats considered in this research are used. Secondly, training programs for lecturers of the Bachelor’s Degree in Primary Education shall be provided, so lecturers will have the means to train their students in the strategic and self-regulated use of the skills involved in written comprehension and production.